MOSU2: Long-range Navigation System with Multi-modal Scene Understanding
cs.RO
Submitted: 2026-11-08
Updated: 2026-11-18
Project page: https://jinggm.github.io/mosu2/Keywords
Terminology
Sources
- AutoSpatial: Visual-Language Reasoning for Social Robot Navigation through Efficient Spatial Reasoning Learning
- PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs
- ViNT: A Foundation Model for Visual Navigation
- Gemini: A Family of Highly Capable Multimodal Models
- SuperNav: An Agentic Navigation System for Any Task in Any Scene
- Walk With Me: Long-Horizon Social Navigation for Human-Centric Outdoor Assistance
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation